setup-matt-pocock-skills
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local repository files, including
AGENTS.md,CLAUDE.md,CONTEXT.md, andpackage.json, to infer the repository's architecture and status. This untrusted data directly informs the setup logic and the configuration drafts presented to the user. - Ingestion points:
SKILL.md(Step 1: Explore) reading from multiple root-level configuration and documentation files. - Boundary markers: Absent; the agent is instructed to read the files directly without explicit delimitation of content.
- Capability inventory: File system read access, execution of
git remote -v, and file system write access for scaffolding documentation files in thedocs/agents/directory. - Sanitization: The skill mitigates risks by requiring the agent to present findings to the user for confirmation (Step 3) before any content is written to the disk.
- [COMMAND_EXECUTION]: The skill instructions specify the execution of the
git remote -vcommand to identify the repository's remote hosting provider (GitHub vs GitLab) during the initial exploration phase.
Audit Metadata